3 static function calculate_article_hash($article, $pluginhost) {
6 foreach ($article as $k => $v) {
7 if ($k != "feed" && isset($v)) {
8 $x = strip_tags(is_array($v) ?
implode(",", $v) : $v);
10 //_debug("$k:" . sha1($x) . ":" . htmlspecialchars($x), true);
12 $tmp .= sha1("$k:" . sha1($x));
16 return sha1(implode(",", $pluginhost->get_plugin_names()) . $tmp);
19 static function update_feedbrowser_cache() {
21 $result = db_query("SELECT feed_url, site_url, title, COUNT(id) AS subscribers
22 FROM ttrss_feeds WHERE feed_url NOT IN (SELECT feed_url FROM ttrss_feeds
23 WHERE private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%')
24 GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000");
28 db_query("DELETE FROM ttrss_feedbrowser_cache");
32 while ($line = db_fetch_assoc($result)) {
33 $subscribers = db_escape_string($line["subscribers"]);
34 $feed_url = db_escape_string($line["feed_url"]);
35 $title = db_escape_string($line["title"]);
36 $site_url = db_escape_string($line["site_url"]);
38 $tmp_result = db_query("SELECT subscribers FROM
39 ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'");
41 if (db_num_rows($tmp_result) == 0) {
43 db_query("INSERT INTO ttrss_feedbrowser_cache
44 (feed_url, site_url, title, subscribers) VALUES ('$feed_url',
45 '$site_url', '$title', '$subscribers')");
59 static function update_daemon_common($limit = DAEMON_FEED_LIMIT
, $debug = true) {
60 $schema_version = get_schema_version();
62 if ($schema_version != SCHEMA_VERSION
) {
63 die("Schema version is wrong, please upgrade the database.\n");
66 if (!SINGLE_USER_MODE
&& DAEMON_UPDATE_LOGIN_LIMIT
> 0) {
67 if (DB_TYPE
== "pgsql") {
68 $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT
." days'";
70 $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT
." DAY)";
73 $login_thresh_qpart = "";
76 if (DB_TYPE
== "pgsql") {
77 $update_limit_qpart = "AND ((
78 ttrss_feeds.update_interval = 0
79 AND ttrss_user_prefs.value != '-1'
80 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL)
82 ttrss_feeds.update_interval > 0
83 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL)
84 ) OR (ttrss_feeds.last_updated IS NULL
85 AND ttrss_user_prefs.value != '-1')
86 OR (last_updated = '1970-01-01 00:00:00'
87 AND ttrss_user_prefs.value != '-1'))";
89 $update_limit_qpart = "AND ((
90 ttrss_feeds.update_interval = 0
91 AND ttrss_user_prefs.value != '-1'
92 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
94 ttrss_feeds.update_interval > 0
95 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
96 ) OR (ttrss_feeds.last_updated IS NULL
97 AND ttrss_user_prefs.value != '-1')
98 OR (last_updated = '1970-01-01 00:00:00'
99 AND ttrss_user_prefs.value != '-1'))";
102 // Test if feed is currently being updated by another process.
103 if (DB_TYPE
== "pgsql") {
104 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')";
106 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 10 MINUTE))";
109 $query_limit = $limit ?
sprintf("LIMIT %d", $limit) : "";
111 // Update the least recently updated feeds first
112 $query_order = "ORDER BY last_updated";
113 if (DB_TYPE
== "pgsql") $query_order .= " NULLS FIRST";
115 $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated
117 ttrss_feeds, ttrss_users, ttrss_user_prefs
119 ttrss_feeds.owner_uid = ttrss_users.id
120 AND ttrss_user_prefs.profile IS NULL
121 AND ttrss_users.id = ttrss_user_prefs.owner_uid
122 AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL'
123 $login_thresh_qpart $update_limit_qpart
124 $updstart_thresh_qpart
125 $query_order $query_limit";
127 $result = db_query($query);
129 if ($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result)));
131 $feeds_to_update = array();
132 while ($line = db_fetch_assoc($result)) {
133 array_push($feeds_to_update, $line['feed_url']);
136 // Update last_update_started before actually starting the batch
137 // in order to minimize collision risk for parallel daemon tasks
138 if (count($feeds_to_update) > 0) {
139 $feeds_quoted = array_map(function ($s) { return "'" . db_escape_string($s) . "'"; }, $feeds_to_update);
141 db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
142 WHERE feed_url IN (%s)", implode(',', $feeds_quoted)));
146 $bstarted = microtime(true);
148 $batch_owners = array();
150 foreach ($feeds_to_update as $feed) {
151 if($debug) _debug("Base feed: $feed");
153 //update_rss_feed($line["id"], true);
155 // since we have the data cached, we can deal with other feeds with the same url
156 $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid
157 FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE
158 ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND
159 ttrss_users.id = ttrss_user_prefs.owner_uid AND
160 ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND
161 ttrss_user_prefs.profile IS NULL AND
162 feed_url = '".db_escape_string($feed)."'
165 ORDER BY ttrss_feeds.id $query_limit");
167 if (db_num_rows($tmp_result) > 0) {
168 while ($tline = db_fetch_assoc($tmp_result)) {
169 if ($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]);
171 if (array_search($tline["owner_uid"], $batch_owners) === FALSE)
172 array_push($batch_owners, $tline["owner_uid"]);
174 $fstarted = microtime(true);
175 RSSUtils
::update_rss_feed($tline["id"], true, false);
176 _debug_suppress(false);
178 _debug(sprintf(" %.4f (sec)", microtime(true) - $fstarted));
186 _debug(sprintf("Processed %d feeds in %.4f (sec), %.4f (sec/feed avg)", $nf,
187 microtime(true) - $bstarted, (microtime(true) - $bstarted) / $nf));
190 foreach ($batch_owners as $owner_uid) {
191 _debug("Running housekeeping tasks for user $owner_uid...");
193 RSSUtils
::housekeeping_user($owner_uid);
196 // Send feed digests by email if needed.
197 Digest
::send_headlines_digests($debug);
203 // this is used when subscribing
204 static function set_basic_feed_info($feed) {
206 $feed = db_escape_string($feed);
208 $result = db_query("SELECT feed_url,auth_pass,auth_login,auth_pass_encrypted
209 FROM ttrss_feeds WHERE id = '$feed'");
211 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
212 0, "auth_pass_encrypted"));
214 $auth_login = db_fetch_result($result, 0, "auth_login");
215 $auth_pass = db_fetch_result($result, 0, "auth_pass");
217 if ($auth_pass_encrypted && function_exists("mcrypt_decrypt")) {
218 require_once "crypt.php";
219 $auth_pass = decrypt_string($auth_pass);
222 $fetch_url = db_fetch_result($result, 0, "feed_url");
224 $feed_data = fetch_file_contents($fetch_url, false,
225 $auth_login, $auth_pass, false,
229 global $fetch_curl_used;
231 if (!$fetch_curl_used) {
232 $tmp = @gzdecode
($feed_data);
234 if ($tmp) $feed_data = $tmp;
237 $feed_data = trim($feed_data);
239 $rss = new FeedParser($feed_data);
242 if (!$rss->error()) {
244 $result = db_query("SELECT title, site_url FROM ttrss_feeds WHERE id = '$feed'");
246 $registered_title = db_fetch_result($result, 0, "title");
247 $orig_site_url = db_fetch_result($result, 0, "site_url");
249 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
250 $feed_title = db_escape_string(mb_substr($rss->get_title(), 0, 199));
252 if ($feed_title && (!$registered_title ||
$registered_title == "[Unknown]")) {
253 db_query("UPDATE ttrss_feeds SET
254 title = '$feed_title' WHERE id = '$feed'");
257 if ($site_url && $orig_site_url != $site_url) {
258 db_query("UPDATE ttrss_feeds SET
259 site_url = '$site_url' WHERE id = '$feed'");
265 * @SuppressWarnings(PHPMD.UnusedFormalParameter)
267 static function update_rss_feed($feed, $no_cache = false) {
269 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') ||
$_REQUEST['xdebug'];
271 _debug_suppress(!$debug_enabled);
272 _debug("start", $debug_enabled);
274 $result = db_query("SELECT title FROM ttrss_feeds
275 WHERE id = '$feed'");
277 if (db_num_rows($result) == 0) {
278 _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled);
279 user_error("Attempt to update unknown/invalid feed $feed", E_USER_WARNING
);
283 $title = db_fetch_result($result, 0, "title");
285 // feed was batch-subscribed or something, we need to get basic info
286 // this is not optimal currently as it fetches stuff separately TODO: optimize
287 if ($title == "[Unknown]") {
288 _debug("setting basic feed info for $feed...");
289 RSSUtils
::set_basic_feed_info($feed);
292 $result = db_query("SELECT id,update_interval,auth_login,
293 feed_url,auth_pass,cache_images,
294 mark_unread_on_update, owner_uid,
295 auth_pass_encrypted, feed_language
296 FROM ttrss_feeds WHERE id = '$feed'");
298 $owner_uid = db_fetch_result($result, 0, "owner_uid");
299 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
300 0, "mark_unread_on_update"));
301 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
302 0, "auth_pass_encrypted"));
304 db_query("UPDATE ttrss_feeds SET last_update_started = NOW()
305 WHERE id = '$feed'");
307 $auth_login = db_fetch_result($result, 0, "auth_login");
308 $auth_pass = db_fetch_result($result, 0, "auth_pass");
310 if ($auth_pass_encrypted && function_exists("mcrypt_decrypt")) {
311 require_once "crypt.php";
312 $auth_pass = decrypt_string($auth_pass);
315 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
316 $fetch_url = db_fetch_result($result, 0, "feed_url");
317 $feed_language = db_escape_string(mb_strtolower(db_fetch_result($result, 0, "feed_language")));
318 if (!$feed_language) $feed_language = 'english';
320 $feed = db_escape_string($feed);
322 $date_feed_processed = date('Y-m-d H:i');
324 $cache_filename = CACHE_DIR
. "/simplepie/" . sha1($fetch_url) . ".xml";
326 $pluginhost = new PluginHost();
327 $pluginhost->set_debug($debug_enabled);
328 $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
330 $pluginhost->load(PLUGINS
, PluginHost
::KIND_ALL
);
331 $pluginhost->load($user_plugins, PluginHost
::KIND_USER
, $owner_uid);
332 $pluginhost->load_data();
336 $force_refetch = isset($_REQUEST["force_refetch"]);
339 foreach ($pluginhost->get_hooks(PluginHost
::HOOK_FETCH_FEED
) as $plugin) {
340 $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, 0, $auth_login, $auth_pass);
345 file_exists($cache_filename) &&
346 is_readable($cache_filename) &&
347 !$auth_login && !$auth_pass &&
348 filemtime($cache_filename) > time() - 30) {
350 _debug("using local cache [$cache_filename].", $debug_enabled);
352 @$feed_data = file_get_contents($cache_filename);
355 $rss_hash = sha1($feed_data);
359 _debug("local cache will not be used for this feed", $debug_enabled);
362 // fetch feed from source
364 _debug("fetching [$fetch_url]...", $debug_enabled);
366 if (ini_get("open_basedir") && function_exists("curl_init")) {
367 _debug("not using CURL due to open_basedir restrictions");
370 $feed_data = fetch_file_contents($fetch_url, false,
371 $auth_login, $auth_pass, false,
372 $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT
: FEED_FETCH_TIMEOUT
,
375 global $fetch_curl_used;
377 if (!$fetch_curl_used) {
378 $tmp = @gzdecode
($feed_data);
380 if ($tmp) $feed_data = $tmp;
383 $feed_data = trim($feed_data);
385 _debug("fetch done.", $debug_enabled);
387 // cache vanilla feed data for re-use
388 if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR
. "/simplepie")) {
389 $new_rss_hash = sha1($feed_data);
391 if ($new_rss_hash != $rss_hash) {
392 _debug("saving $cache_filename", $debug_enabled);
393 @file_put_contents
($cache_filename, $feed_data);
399 global $fetch_last_error;
400 global $fetch_last_error_code;
402 _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled);
407 if ($fetch_last_error_code != 304) {
408 $error_escaped = db_escape_string($fetch_last_error);
410 _debug("source claims data not modified, nothing to do.", $debug_enabled);
414 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
415 last_updated = NOW() WHERE id = '$feed'");
420 foreach ($pluginhost->get_hooks(PluginHost
::HOOK_FEED_FETCHED
) as $plugin) {
421 $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
424 $rss = new FeedParser($feed_data);
427 $feed = db_escape_string($feed);
429 if (!$rss->error()) {
431 // We use local pluginhost here because we need to load different per-user feed plugins
432 $pluginhost->run_hooks(PluginHost
::HOOK_FEED_PARSED
, "hook_feed_parsed", $rss);
434 _debug("language: $feed_language", $debug_enabled);
435 _debug("processing feed data...", $debug_enabled);
437 // db_query("BEGIN");
439 if (DB_TYPE
== "pgsql") {
440 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
442 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
445 $result = db_query("SELECT owner_uid,favicon_avg_color,
446 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
448 FROM ttrss_feeds WHERE id = '$feed'");
450 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
451 "favicon_needs_check"));
452 $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
454 $owner_uid = db_fetch_result($result, 0, "owner_uid");
456 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
458 _debug("site_url: $site_url", $debug_enabled);
459 _debug("feed_title: " . $rss->get_title(), $debug_enabled);
461 if ($favicon_needs_check ||
$force_refetch) {
463 /* terrible hack: if we crash on floicon shit here, we won't check
464 * the icon avgcolor again (unless the icon got updated) */
466 $favicon_file = ICONS_DIR
. "/$feed.ico";
467 $favicon_modified = @filemtime
($favicon_file);
469 _debug("checking favicon...", $debug_enabled);
471 RSSUtils
::check_feed_favicon($site_url, $feed);
472 $favicon_modified_new = @filemtime
($favicon_file);
474 if ($favicon_modified_new > $favicon_modified)
475 $favicon_avg_color = '';
477 if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
478 require_once "colors.php";
480 db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE
483 $favicon_color = db_escape_string(
484 calculate_avg_color($favicon_file));
486 $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'";
487 } else if ($favicon_avg_color == 'fail') {
488 _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
491 db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()
493 WHERE id = '$feed'");
496 _debug("loading filters & labels...", $debug_enabled);
498 $filters = load_filters($feed, $owner_uid);
500 if ($debug_enabled) {
504 _debug("" . count($filters) . " filters loaded.", $debug_enabled);
506 $items = $rss->get_items();
508 if (!is_array($items)) {
509 _debug("no articles found.", $debug_enabled);
511 db_query("UPDATE ttrss_feeds
512 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
514 return; // no articles
517 _debug("processing articles...", $debug_enabled);
521 foreach ($items as $item) {
522 if ($_REQUEST['xdebug'] == 3) {
526 if (ini_get("max_execution_time") > 0 && time() - $tstart >= ini_get("max_execution_time") * 0.7) {
527 _debug("looks like there's too many articles to process at once, breaking out", $debug_enabled);
531 $entry_guid = $item->get_id();
532 if (!$entry_guid) $entry_guid = $item->get_link();
533 if (!$entry_guid) $entry_guid = RSSUtils
::make_guid_from_title($item->get_title());
534 if (!$entry_guid) continue;
536 $entry_guid = "$owner_uid,$entry_guid";
538 $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
540 _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled);
542 $entry_timestamp = "";
544 $entry_timestamp = $item->get_date();
546 _debug("orig date: " . $item->get_date(), $debug_enabled);
548 if ($entry_timestamp == -1 ||
!$entry_timestamp ||
$entry_timestamp > time()) {
549 $entry_timestamp = time();
552 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
554 _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled);
556 // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
557 // $entry_title = decode_numeric_entities($entry_title);
558 $entry_title = $item->get_title();
560 $entry_link = rewrite_relative_url($site_url, $item->get_link());
562 _debug("title $entry_title", $debug_enabled);
563 _debug("link $entry_link", $debug_enabled);
565 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
567 $entry_content = $item->get_content();
568 if (!$entry_content) $entry_content = $item->get_description();
570 if ($_REQUEST["xdebug"] == 2) {
572 print htmlspecialchars($entry_content);
576 $entry_comments = db_escape_string(mb_substr($item->get_comments_url(), 0, 245));
577 $num_comments = (int) $item->get_comments_count();
579 $entry_author = $item->get_author(); // escaped later
580 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
582 _debug("author $entry_author", $debug_enabled);
583 _debug("num_comments: $num_comments", $debug_enabled);
584 _debug("looking for tags...", $debug_enabled);
586 // parse <category> entries into tags
588 $additional_tags = array();
590 $additional_tags_src = $item->get_categories();
592 if (is_array($additional_tags_src)) {
593 foreach ($additional_tags_src as $tobj) {
594 array_push($additional_tags, $tobj);
598 $entry_tags = array_unique($additional_tags);
600 for ($i = 0; $i < count($entry_tags); $i++
)
601 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
603 _debug("tags found: " . join(",", $entry_tags), $debug_enabled);
605 _debug("done collecting data.", $debug_enabled);
607 $result = db_query("SELECT id, content_hash, lang FROM ttrss_entries
608 WHERE guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed'");
610 if (db_num_rows($result) != 0) {
611 $base_entry_id = db_fetch_result($result, 0, "id");
612 $entry_stored_hash = db_fetch_result($result, 0, "content_hash");
613 $article_labels = Article
::get_article_labels($base_entry_id, $owner_uid);
614 $entry_language = db_fetch_result($result, 0, "lang");
616 $existing_tags = Article
::get_article_tags($base_entry_id, $owner_uid);
617 $entry_tags = array_unique(array_merge($entry_tags, $existing_tags));
620 $base_entry_id = false;
621 $entry_stored_hash = "";
622 $article_labels = array();
623 $entry_language = "";
626 $article = array("owner_uid" => $owner_uid, // read only
627 "guid" => $entry_guid, // read only
628 "guid_hashed" => $entry_guid_hashed, // read only
629 "title" => $entry_title,
630 "content" => $entry_content,
631 "link" => $entry_link,
632 "labels" => $article_labels, // current limitation: can add labels to article, can't remove them
633 "tags" => $entry_tags,
634 "author" => $entry_author,
635 "force_catchup" => false, // ugly hack for the time being
636 "score_modifier" => 0, // no previous value, plugin should recalculate score modifier based on content if needed
637 "language" => $entry_language,
638 "feed" => array("id" => $feed,
639 "fetch_url" => $fetch_url,
640 "site_url" => $site_url,
641 "cache_images" => $cache_images)
644 $entry_plugin_data = "";
645 $entry_current_hash = RSSUtils
::calculate_article_hash($article, $pluginhost);
647 _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled);
649 if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) {
650 _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled);
652 // we keep encountering the entry in feeds, so we need to
653 // update date_updated column so that we don't get horrible
654 // dupes when the entry gets purged and reinserted again e.g.
655 // in the case of SLOW SLOW OMG SLOW updating feeds
657 $base_entry_id = db_fetch_result($result, 0, "id");
659 db_query("UPDATE ttrss_entries SET date_updated = NOW()
660 WHERE id = '$base_entry_id'");
665 _debug("hash differs, applying plugin filters:", $debug_enabled);
667 foreach ($pluginhost->get_hooks(PluginHost
::HOOK_ARTICLE_FILTER
) as $plugin) {
668 _debug("... " . get_class($plugin), $debug_enabled);
670 $start = microtime(true);
671 $article = $plugin->hook_article_filter($article);
673 _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled);
675 $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
678 if ($_REQUEST["xdebug"] == 2) {
679 print "processed content: ";
680 print htmlspecialchars($article["content"]);
684 $entry_plugin_data = db_escape_string($entry_plugin_data);
686 _debug("plugin data: $entry_plugin_data", $debug_enabled);
688 // Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077
689 if (DB_TYPE
== "mysql") {
690 foreach ($article as $k => $v) {
692 // i guess we'll have to take the risk of 4byte unicode labels & tags here
693 if (is_string($article[$k])) {
694 $article[$k] = preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $v);
699 /* Collect article tags here so we could filter by them: */
701 $matched_rules = array();
703 $article_filters = RSSUtils
::get_article_filters($filters, $article["title"],
704 $article["content"], $article["link"], $article["author"],
705 $article["tags"], $matched_rules);
707 if ($debug_enabled) {
708 _debug("matched filter rules: ", $debug_enabled);
710 if (count($matched_rules) != 0) {
711 print_r($matched_rules);
714 _debug("filter actions: ", $debug_enabled);
716 if (count($article_filters) != 0) {
717 print_r($article_filters);
721 $plugin_filter_names = RSSUtils
::find_article_filters($article_filters, "plugin");
722 $plugin_filter_actions = $pluginhost->get_filter_actions();
724 if (count($plugin_filter_names) > 0) {
725 _debug("applying plugin filter actions...", $debug_enabled);
727 foreach ($plugin_filter_names as $pfn) {
728 list($pfclass,$pfaction) = explode(":", $pfn["param"]);
730 if (isset($plugin_filter_actions[$pfclass])) {
731 $plugin = $pluginhost->get_plugin($pfclass);
733 _debug("... $pfclass: $pfaction", $debug_enabled);
736 $start = microtime(true);
737 $article = $plugin->hook_article_filter_action($article, $pfaction);
739 _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled);
741 _debug("??? $pfclass: plugin object not found.");
744 _debug("??? $pfclass: filter plugin not registered.");
749 $entry_tags = $article["tags"];
750 $entry_guid = db_escape_string($entry_guid);
751 $entry_title = db_escape_string($article["title"]);
752 $entry_author = db_escape_string(mb_substr($article["author"], 0, 245));
753 $entry_link = db_escape_string($article["link"]);
754 $entry_content = $article["content"]; // escaped below
755 $entry_force_catchup = $article["force_catchup"];
756 $article_labels = $article["labels"];
757 $entry_score_modifier = (int) $article["score_modifier"];
758 $entry_language = db_escape_string($article["language"]);
760 if ($debug_enabled) {
761 _debug("article labels:", $debug_enabled);
763 if (count($article_labels) != 0) {
764 print_r($article_labels);
768 _debug("force catchup: $entry_force_catchup");
770 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
771 RSSUtils
::cache_media($entry_content, $site_url, $debug_enabled);
773 $entry_content = db_escape_string($entry_content, false);
777 $result = db_query("SELECT id FROM ttrss_entries
778 WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')");
780 if (db_num_rows($result) == 0) {
782 _debug("base guid [$entry_guid or $entry_guid_hashed] not found, creating...", $debug_enabled);
784 // base post entry does not exist, create it
787 "INSERT INTO ttrss_entries
804 '$entry_guid_hashed',
806 '$entry_timestamp_fmt',
808 '$entry_current_hash',
811 '$date_feed_processed',
814 '$entry_plugin_data',
820 // now it should exist, if not - bad luck then
822 $result = db_query("SELECT id FROM ttrss_entries
823 WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'");
828 if (db_num_rows($result) == 1) {
830 _debug("base guid found, checking for user record", $debug_enabled);
832 $ref_id = db_fetch_result($result, 0, "id");
833 $entry_ref_id = $ref_id;
835 /* $stored_guid = db_fetch_result($result, 0, "guid");
836 if ($stored_guid != $entry_guid_hashed) {
837 if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
839 db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
843 if (RSSUtils
::find_article_filter($article_filters, "filter")) {
844 //db_query("COMMIT"); // close transaction in progress
848 $score = RSSUtils
::calculate_article_score($article_filters) +
$entry_score_modifier;
850 _debug("initial score: $score [including plugin modifier: $entry_score_modifier]", $debug_enabled);
852 // check for user post link to main table
854 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
855 ref_id = '$ref_id' AND owner_uid = '$owner_uid'";
857 // if ($_REQUEST["xdebug"]) print "$query\n";
859 $result = db_query($query);
861 // okay it doesn't exist - create user entry
862 if (db_num_rows($result) == 0) {
864 _debug("user record not found, creating...", $debug_enabled);
866 if ($score >= -500 && !RSSUtils
::find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) {
868 $last_read_qpart = 'NULL';
871 $last_read_qpart = 'NOW()';
874 if (RSSUtils
::find_article_filter($article_filters, 'mark') ||
$score > 1000) {
880 if (RSSUtils
::find_article_filter($article_filters, 'publish')) {
883 $published = 'false';
886 $last_marked = ($marked == 'true') ?
'NOW()' : 'NULL';
887 $last_published = ($published == 'true') ?
'NOW()' : 'NULL';
890 "INSERT INTO ttrss_user_entries
891 (ref_id, owner_uid, feed_id, unread, last_read, marked,
892 published, score, tag_cache, label_cache, uuid,
893 last_marked, last_published)
894 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
895 $last_read_qpart, $marked, $published, '$score', '', '',
896 '', $last_marked, $last_published)");
899 "SELECT int_id FROM ttrss_user_entries WHERE
900 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
901 feed_id = '$feed' LIMIT 1");
903 if (db_num_rows($result) == 1) {
904 $entry_int_id = db_fetch_result($result, 0, "int_id");
907 _debug("user record FOUND", $debug_enabled);
909 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
910 $entry_int_id = db_fetch_result($result, 0, "int_id");
913 _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled);
915 if (DB_TYPE
== "pgsql") {
916 $tsvector_combined = db_escape_string(mb_substr($entry_title . ' ' . strip_tags(str_replace('<', ' <', $entry_content)),
919 $tsvector_qpart = "tsvector_combined = to_tsvector('$feed_language', '$tsvector_combined'),";
922 $tsvector_qpart = "";
925 db_query("UPDATE ttrss_entries
926 SET title = '$entry_title',
927 content = '$entry_content',
928 content_hash = '$entry_current_hash',
929 updated = '$entry_timestamp_fmt',
931 num_comments = '$num_comments',
932 plugin_data = '$entry_plugin_data',
933 author = '$entry_author',
934 lang = '$entry_language'
935 WHERE id = '$ref_id'");
938 db_query("UPDATE ttrss_user_entries
939 SET score = '$score' WHERE ref_id = '$ref_id'");
941 if ($mark_unread_on_update) {
942 _debug("article updated, marking unread as requested.", $debug_enabled);
944 db_query("UPDATE ttrss_user_entries
945 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
949 //db_query("COMMIT");
951 _debug("assigning labels [other]...", $debug_enabled);
953 foreach ($article_labels as $label) {
954 Labels
::add_article($entry_ref_id, $label[1], $owner_uid);
957 _debug("assigning labels [filters]...", $debug_enabled);
959 RSSUtils
::assign_article_to_label_filters($entry_ref_id, $article_filters,
960 $owner_uid, $article_labels);
962 _debug("looking for enclosures...", $debug_enabled);
966 $enclosures = array();
968 $encs = $item->get_enclosures();
970 if (is_array($encs)) {
971 foreach ($encs as $e) {
973 rewrite_relative_url($site_url, $e->link
),
974 $e->type
, $e->length
, $e->title
, $e->width
, $e->height
);
975 array_push($enclosures, $e_item);
979 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
980 RSSUtils
::cache_enclosures($enclosures, $site_url, $debug_enabled);
982 if ($debug_enabled) {
983 _debug("article enclosures:", $debug_enabled);
984 print_r($enclosures);
990 // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
992 foreach ($enclosures as $enc) {
993 $enc_url = db_escape_string($enc[0]);
994 $enc_type = db_escape_string($enc[1]);
995 $enc_dur = db_escape_string($enc[2]);
996 $enc_title = db_escape_string($enc[3]);
997 $enc_width = intval($enc[4]);
998 $enc_height = intval($enc[5]);
1000 $result = db_query("SELECT id FROM ttrss_enclosures
1001 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1003 if (db_num_rows($result) == 0) {
1004 db_query("INSERT INTO ttrss_enclosures
1005 (content_url, content_type, title, duration, post_id, width, height) VALUES
1006 ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)");
1010 //db_query("COMMIT");
1012 // check for manual tags (we have to do it here since they're loaded from filters)
1014 foreach ($article_filters as $f) {
1015 if ($f["type"] == "tag") {
1017 $manual_tags = trim_array(explode(",", $f["param"]));
1019 foreach ($manual_tags as $tag) {
1020 if (tag_is_valid($tag)) {
1021 array_push($entry_tags, $tag);
1029 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref(
1030 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1032 $filtered_tags = array();
1033 $tags_to_cache = array();
1035 if ($entry_tags && is_array($entry_tags)) {
1036 foreach ($entry_tags as $tag) {
1037 if (array_search($tag, $boring_tags) === false) {
1038 array_push($filtered_tags, $tag);
1043 $filtered_tags = array_unique($filtered_tags);
1045 if ($debug_enabled) {
1046 _debug("filtered article tags:", $debug_enabled);
1047 print_r($filtered_tags);
1050 // Save article tags in the database
1052 if (count($filtered_tags) > 0) {
1054 //db_query("BEGIN");
1056 foreach ($filtered_tags as $tag) {
1058 $tag = sanitize_tag($tag);
1059 $tag = db_escape_string($tag);
1061 if (!tag_is_valid($tag)) continue;
1063 $result = db_query("SELECT id FROM ttrss_tags
1064 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1065 owner_uid = '$owner_uid' LIMIT 1");
1067 if ($result && db_num_rows($result) == 0) {
1069 db_query("INSERT INTO ttrss_tags
1070 (owner_uid,tag_name,post_int_id)
1071 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1074 array_push($tags_to_cache, $tag);
1077 /* update the cache */
1079 $tags_to_cache = array_unique($tags_to_cache);
1081 $tags_str = db_escape_string(join(",", $tags_to_cache));
1083 db_query("UPDATE ttrss_user_entries
1084 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1085 AND owner_uid = $owner_uid");
1087 //db_query("COMMIT");
1090 _debug("article processed", $debug_enabled);
1093 _debug("purging feed...", $debug_enabled);
1095 purge_feed($feed, 0, $debug_enabled);
1097 db_query("UPDATE ttrss_feeds
1098 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1100 // db_query("COMMIT");
1104 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
1106 _debug("fetch error: $error_msg", $debug_enabled);
1108 if (count($rss->errors()) > 1) {
1109 foreach ($rss->errors() as $error) {
1115 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1116 last_updated = NOW() WHERE id = '$feed'");
1122 _debug("done", $debug_enabled);
1127 static function cache_enclosures($enclosures, $site_url, $debug) {
1128 foreach ($enclosures as $enc) {
1130 if (preg_match("/(image|audio|video)/", $enc[1])) {
1132 $src = rewrite_relative_url($site_url, $enc[0]);
1134 $local_filename = CACHE_DIR
. "/images/" . sha1($src);
1136 if ($debug) _debug("cache_enclosures: downloading: $src to $local_filename");
1138 if (!file_exists($local_filename)) {
1139 $file_content = fetch_file_contents($src);
1141 if ($file_content && strlen($file_content) > MIN_CACHE_FILE_SIZE
) {
1142 file_put_contents($local_filename, $file_content);
1145 touch($local_filename);
1151 static function cache_media($html, $site_url, $debug) {
1152 libxml_use_internal_errors(true);
1154 $charset_hack = '<head>
1155 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1158 $doc = new DOMDocument();
1159 $doc->loadHTML($charset_hack . $html);
1160 $xpath = new DOMXPath($doc);
1162 $entries = $xpath->query('(//img[@src])|(//video/source[@src])|(//audio/source[@src])');
1164 foreach ($entries as $entry) {
1165 if ($entry->hasAttribute('src') && strpos($entry->getAttribute('src'), "data:") !== 0) {
1166 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1168 $local_filename = CACHE_DIR
. "/images/" . sha1($src);
1170 if ($debug) _debug("cache_media: downloading: $src to $local_filename");
1172 if (!file_exists($local_filename)) {
1173 $file_content = fetch_file_contents($src);
1175 if ($file_content && strlen($file_content) > MIN_CACHE_FILE_SIZE
) {
1176 file_put_contents($local_filename, $file_content);
1179 touch($local_filename);
1185 static function expire_error_log($debug) {
1186 if ($debug) _debug("Removing old error log entries...");
1188 if (DB_TYPE
== "pgsql") {
1189 db_query("DELETE FROM ttrss_error_log
1190 WHERE created_at < NOW() - INTERVAL '7 days'");
1192 db_query("DELETE FROM ttrss_error_log
1193 WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)");
1198 static function expire_lock_files($debug) {
1199 //if ($debug) _debug("Removing old lock files...");
1203 if (is_writable(LOCK_DIRECTORY
)) {
1204 $files = glob(LOCK_DIRECTORY
. "/*.lock");
1207 foreach ($files as $file) {
1208 if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) {
1216 if ($debug) _debug("Removed $num_deleted old lock files.");
1219 static function expire_cached_files($debug) {
1220 foreach (array("simplepie", "images", "export", "upload") as $dir) {
1221 $cache_dir = CACHE_DIR
. "/$dir";
1223 // if ($debug) _debug("Expiring $cache_dir");
1227 if (is_writable($cache_dir)) {
1228 $files = glob("$cache_dir/*");
1231 foreach ($files as $file) {
1232 if (time() - filemtime($file) > 86400*CACHE_MAX_DAYS
) {
1241 if ($debug) _debug("$cache_dir: removed $num_deleted files.");
1246 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1247 * Returns the url query as associative array
1249 * @param string query
1250 * @return array params
1252 static function convertUrlQuery($query) {
1253 $queryParts = explode('&', $query);
1257 foreach ($queryParts as $param) {
1258 $item = explode('=', $param);
1259 $params[$item[0]] = $item[1];
1265 static function get_article_filters($filters, $title, $content, $link, $author, $tags, &$matched_rules = false) {
1268 foreach ($filters as $filter) {
1269 $match_any_rule = $filter["match_any_rule"];
1270 $inverse = $filter["inverse"];
1271 $filter_match = false;
1273 foreach ($filter["rules"] as $rule) {
1275 $reg_exp = str_replace('/', '\/', $rule["reg_exp"]);
1276 $rule_inverse = $rule["inverse"];
1281 switch ($rule["type"]) {
1283 $match = @preg_match
("/$reg_exp/iu", $title);
1286 // we don't need to deal with multiline regexps
1287 $content = preg_replace("/[\r\n\t]/", "", $content);
1289 $match = @preg_match
("/$reg_exp/iu", $content);
1292 // we don't need to deal with multiline regexps
1293 $content = preg_replace("/[\r\n\t]/", "", $content);
1295 $match = (@preg_match
("/$reg_exp/iu", $title) || @preg_match
("/$reg_exp/iu", $content));
1298 $match = @preg_match
("/$reg_exp/iu", $link);
1301 $match = @preg_match
("/$reg_exp/iu", $author);
1304 foreach ($tags as $tag) {
1305 if (@preg_match
("/$reg_exp/iu", $tag)) {
1313 if ($rule_inverse) $match = !$match;
1315 if ($match_any_rule) {
1317 $filter_match = true;
1321 $filter_match = $match;
1328 if ($inverse) $filter_match = !$filter_match;
1330 if ($filter_match) {
1331 if (is_array($matched_rules)) array_push($matched_rules, $rule);
1333 foreach ($filter["actions"] AS $action) {
1334 array_push($matches, $action);
1336 // if Stop action encountered, perform no further processing
1337 if (isset($action["type"]) && $action["type"] == "stop") return $matches;
1345 static function find_article_filter($filters, $filter_name) {
1346 foreach ($filters as $f) {
1347 if ($f["type"] == $filter_name) {
1354 static function find_article_filters($filters, $filter_name) {
1357 foreach ($filters as $f) {
1358 if ($f["type"] == $filter_name) {
1359 array_push($results, $f);
1365 static function calculate_article_score($filters) {
1368 foreach ($filters as $f) {
1369 if ($f["type"] == "score") {
1370 $score +
= $f["param"];
1376 static function labels_contains_caption($labels, $caption) {
1377 foreach ($labels as $label) {
1378 if ($label[1] == $caption) {
1386 static function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) {
1387 foreach ($filters as $f) {
1388 if ($f["type"] == "label") {
1389 if (!RSSUtils
::labels_contains_caption($article_labels, $f["param"])) {
1390 Labels
::add_article($id, $f["param"], $owner_uid);
1396 static function make_guid_from_title($title) {
1397 return preg_replace("/[ \"\',.:;]/", "-",
1398 mb_strtolower(strip_tags($title), 'utf-8'));
1401 static function cleanup_counters_cache($debug) {
1402 $result = db_query("DELETE FROM ttrss_counters_cache
1403 WHERE feed_id > 0 AND
1404 (SELECT COUNT(id) FROM ttrss_feeds WHERE
1406 ttrss_counters_cache.owner_uid = ttrss_feeds.owner_uid) = 0");
1407 $frows = db_affected_rows($result);
1409 $result = db_query("DELETE FROM ttrss_cat_counters_cache
1410 WHERE feed_id > 0 AND
1411 (SELECT COUNT(id) FROM ttrss_feed_categories WHERE
1413 ttrss_cat_counters_cache.owner_uid = ttrss_feed_categories.owner_uid) = 0");
1414 $crows = db_affected_rows($result);
1416 if ($debug) _debug("Removed $frows (feeds) $crows (cats) orphaned counter cache entries.");
1419 static function housekeeping_user($owner_uid) {
1420 $tmph = new PluginHost();
1422 load_user_plugins($owner_uid, $tmph);
1424 $tmph->run_hooks(PluginHost
::HOOK_HOUSE_KEEPING
, "hook_house_keeping", "");
1427 static function housekeeping_common($debug) {
1428 RSSUtils
::expire_cached_files($debug);
1429 RSSUtils
::expire_lock_files($debug);
1430 RSSUtils
::expire_error_log($debug);
1432 $count = RSSUtils
::update_feedbrowser_cache();
1433 _debug("Feedbrowser updated, $count feeds processed.");
1435 Article
::purge_orphans( true);
1436 RSSUtils
::cleanup_counters_cache($debug);
1438 //$rc = cleanup_tags( 14, 50000);
1439 //_debug("Cleaned $rc cached tags.");
1441 PluginHost
::getInstance()->run_hooks(PluginHost
::HOOK_HOUSE_KEEPING
, "hook_house_keeping", "");
1444 static function check_feed_favicon($site_url, $feed) {
1445 # print "FAVICON [$site_url]: $favicon_url\n";
1447 $icon_file = ICONS_DIR
. "/$feed.ico";
1449 if (!file_exists($icon_file)) {
1450 $favicon_url = get_favicon_url($site_url);
1453 // Limiting to "image" type misses those served with text/plain
1454 $contents = fetch_file_contents($favicon_url); // , "image");
1457 // Crude image type matching.
1458 // Patterns gleaned from the file(1) source code.
1459 if (preg_match('/^\x00\x00\x01\x00/', $contents)) {
1460 // 0 string \000\000\001\000 MS Windows icon resource
1461 //error_log("check_feed_favicon: favicon_url=$favicon_url isa MS Windows icon resource");
1463 elseif (preg_match('/^GIF8/', $contents)) {
1464 // 0 string GIF8 GIF image data
1465 //error_log("check_feed_favicon: favicon_url=$favicon_url isa GIF image");
1467 elseif (preg_match('/^\x89PNG\x0d\x0a\x1a\x0a/', $contents)) {
1468 // 0 string \x89PNG\x0d\x0a\x1a\x0a PNG image data
1469 //error_log("check_feed_favicon: favicon_url=$favicon_url isa PNG image");
1471 elseif (preg_match('/^\xff\xd8/', $contents)) {
1472 // 0 beshort 0xffd8 JPEG image data
1473 //error_log("check_feed_favicon: favicon_url=$favicon_url isa JPG image");
1476 //error_log("check_feed_favicon: favicon_url=$favicon_url isa UNKNOWN type");
1482 $fp = @fopen
($icon_file, "w");
1485 fwrite($fp, $contents);
1487 chmod($icon_file, 0644);